Molecular characterization of a novel polymycovirus from the phytopathogenic fungus Setosphaeria turcica

A putative polymycovirus tentatively named "Setosphaeria turcica polymycovirus 1" (StPmV1) was discovered in the phytopathogenic fungus Setosphaeria turcica. StPmV1 has a genome comprising five double-stranded RNAs (dsRNAs). dsRNA1, 2, and 3 each encode a protein sharing significant similarity but lower than 64% sequence identity to the corresponding proteins of other polymycoviruses. dsRNA4 and 5 each encode a protein with a sequence that is not conserved among polymycoviruses. However, the protein encoded by dsRNA4 is rich in proline (P), alanine (A), and serine (S) residues, which is a feature shared by the so-called PAS-rich proteins encoded by all polymycoviruses. Phylogeny reconstruction using the RNA-dependent RNA polymerase (RdRp) sequences of accepted or putative polymycoviruses revealed that StPmV1 is most closely related to Plasmopara viticola lesion associated polymycovirus 1 (PvaPolymyco1), a putative polymycovirus recovered from the phytopathogenic oomycetes Plasmopara viticola. These data suggest that StPmV1 may represent a novel species of the genus Polymycovirus, family Polymycoviridae. To our knowledge, this is the first polymycovirus reported from S. turcica.


Provenance of the virus material
S. turcica strain ZY17 was isolated from a corn leaf showing northern leaf blight symptoms from Zaoyang, Hubei province, China, in 2019. After growth for 6 days on a potato dextrose agar (PDA) plate at 28°C, mycelia (~0.2 g) of ZY17 were harvested and subjected to dsRNA extraction using a procedure described previously [13]. The dsRNA molecules in the extract were separated by electrophoresis in a 1% agarose gel. Four discrete bands were observed in the gel after ethidium bromide (EB) staining (Fig. 1A). Assuming that dsRNAs have a migration speed comparable to those of dsDNAs, this banding pattern suggests the presence of multiple dsRNA elements ranging in size between 1 and 2.5 kb. To sequence these dsRNAs, each band was excised from the gel. dsRNA from each gel slice was recovered using a TIANgel Midi Purification Kit (Tiangen, China) and reverse transcribed using the tagged random primer 5'-CGA TCG ATC ATG ATG CAA TGCNNNNNN-3'. The resulting complementary DNA (cDNA) was amplified using the primer 5'-CGA TCG ATC ATG ATG CAA TGC-3'. PCR amplicons longer than 500 bp were each ligated to the vector pMD-20T (TAKARA), and the recombinant plasmids were introduced into Escherichia coli DH5a by transformation. For each transformation, 8-14 randomly selected colonies were sent to Beijing Genomics Institute (BGI, China) for Sanger sequencing. The sequences were analyzed and assembled using DNASTAR. The 5'-and 3'-terminal sequences of each dsRNA were determined by RNA-ligase-mediated rapid amplification of cDNA ends (RLM-RACE) [14]. The results showed that the smallest band was formed by two distinct dsRNAs, 1179 bp and 1115 bp in length. Each of the three larger bands represents a single dsRNA with a size of 2427, 2259, and 1994 bp, respectively. The five dsRNAs are named dsRNA1-5 in decreasing order of size, and their sequences were deposited in the GenBank database under the accession numbers MW429374.1-MW429378.1. Each nucleotide position on each dsRNA was sequenced at least four times.
P2 has a molecular mass of 74.96 kDa. No conserved domain was found in P2. However, P2 shows 29.48%-47.49% sequence identity to dsRNA2-encoded proteins from many, although not all, polymycoviruses. The N-terminus of P2 has the sequence MADLTRL, which, as part of a signal peptide, is found at the N-termini of dsRNA2encoded proteins of all polymycoviruses, as noted by Kotta-Loizou and Coutts, who proposed that polymycoviruses may use the protein encoded by dsRNA2 to anchor their replication machinery to a membranous compartment within the host cell [4]. Supporting this, a transmembrane helix from amino acid residue 279 to residue 298, was strongly predicted in P2 using the TMpred server (data not shown) [20]. Notably, a hypothetical protein encoded by an unnamed dsRNA (GenBank accession no. ACL80752) from a fungus of the genus Alternaria showed 60.34% amino acid sequence identity to P2.
P3 has a molecular mass of 65.68 kDa and is predicted to contain a methyltransferase domain (pfam13649) spanning amino acid residue position 131 to position 236. This methyltransferase domain has been detected in the dsRNA3-encoded proteins of all polymycoviruses [1][2][3][4][5][6][7][8][9]. Although experimental data are lacking at present, the presence of this domain suggests that polymycoviruses may use dsRNA3-encoded proteins to cap their genomic or messenger RNAs [1]. Like P2, P3 shows significant similarity to cognate proteins from a subset of polymycoviruses. The most similar protein, encoded by dsRNA3 of PvaPolymyco1, shares 59.54% amino acid sequence identity with P3 [9].
P5 has a molecular mass of 33.81 kDa. This protein does not show similarity to any known proteins, nor does it contain any conserved domains. Thus, the function of P5 cannot be predicted at present.
To investigate the relationships between StPmV1 and other polymycoviruses, the RdRp sequences of accepted or putative polymycoviruses were used for phylogeny reconstruction with IQ-TREE [22,23]. The RdRp of Hadaka virus 1 (HadV1), a polymycovirus-related virus, was used as an outgroup. The maximum-likelihood tree returned from IQ-TREE divided polymycoviruses into four major clusters, although the relationships between these clusters are not well resolved (Fig. 2B). StPmV1 was placed in cluster 4. As a comparison, AfuTmV-1, the type member of the family Polymycoviridae, was placed in cluster 3 [1]. CcFV1, the only polymycovirus known to form filamentous virions, was placed in cluster 2 [3]. Beauveria bassiana polymycovirus 1 (BbPmV1), a polymycovirus infecting the entomopathogenic fungus Beauveria bassiana, was placed in cluster 1. Within cluster 4, StPmV1 formed a monophyletic branch with PvaPolymyco1. Interestingly, PvaPolymyco3 and PvaPolymyco5, which were recovered from the same host as PvaPolymyco1, were not included in this branch [9].
Overall, StPmV1 has a genome with features characteristic of polymycoviruses. However, StPmV1 encodes an RdRp sharing <64% sequence identity with the RdRp of other polymycoviruses. StPmV1 may infect S. turcica, a fungus from which no polymycovirus has been reported. These data suggest that StPmV1 represents a novel species of the genus Polymycovirus, family Polymycoviridae. Because some polymycoviruses have been shown to affect the pathogenicity of their hosts, further studies on the effects of StPmV1 on S. turcica would be of interest.